Skip to content

The project covers the deficit that exist in the online available resources in data science for beginners, whereby the existing models work just fine on well prepared mnist, fashion-mnist etc datasets but often students find it difficult to achieve the kind of accuracy for their own datasets. The project intends to cover simple methods to genera…

Notifications You must be signed in to change notification settings

chahalinder0007/preprocessing-images-from-scratch

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

preprocessing-images-from-scratch

The project covers the deficit that exist in the online available resources in data science for beginners, whereby the existing models work just fine on well prepared mnist, fashion-mnist etc datasets but often students find it difficult to achieve the kind of accuracy for their own datasets. The project intends to cover simple methods to generate a good dataset from different images obtained from any sources.

Requirements: although the requirements have been taken care of as the script will initialize and install the required dependencies but here is a list anyway:

  • scipy
  • opencv
  • sklearn
  • pywt
  • numpy
  • math

The best suggested method is to use a jupyter notebook to open the mainn.ipynb file. it will execute all the required files. The files are assumed to be mnist type where each image is placed in a directory which is its label.

P.S. You are free to use the entire project in whatever way you deem fit, you must do your own due diligence before implementing it for any comercial implementation as I will not be liable for any defects, bugs etc that may cause any financial or any other kind of loss to the extent permitted by law. The project utilises popular open-source liabraries.

About

The project covers the deficit that exist in the online available resources in data science for beginners, whereby the existing models work just fine on well prepared mnist, fashion-mnist etc datasets but often students find it difficult to achieve the kind of accuracy for their own datasets. The project intends to cover simple methods to genera…

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published